Summary
The 2023 Access to Credit Report is a research report that evaluates Upstart's personal loan underwriting model in comparison to a traditional model. The report outlines the findings from the analysis conducted over the previous year, highlighting that Upstart's AI model approved significantly more applicants, including those from Black and Hispanic backgrounds, while offering lower average annual percentage rates (APRs). Specifically, the Upstart model approved 101% more applicants overall, with 116% more Black applicants and 123% more Hispanic applicants compared to the traditional model. The report details the construction of the traditional model, which employs a Cox Proportional Hazard modeling framework with seven specific inputs. It also describes the methodology used to simulate approval and pricing decisions for applicants in 2023, ensuring a fair comparison between the two models. The report aims to enhance understanding of Upstart's efforts to improve access to affordable and inclusive credit across the nation.
